During 1978 and 1979 two infant formulas deficient in chloride were marketed in the United States. It has been estimated that a minimum of 20,000 infant years of these formulas were purchased and more than 100 children were reported to the Centers for Disease Control with metabolic and other abnormalities, principally hypochloremic metabolic alkalosis. In a study of 21 of these children at 2 years of age a significant inverse correlation between length of exclusive use of defective formula and cognitive outcome as measured by the Bayley Scales of Infant Development (r = -.55, p = .01) was noted. In a population based study which ascertained the infant formulas used by first and second graders attending public school those who were exposed to defective formula scored lower on the general cognitive index and the quantitative scale (McCarthy) than did the children who used other soy formulas. To substantiate these findings a further study of children is in progress in the metropolitan Washington. D.C. area schools. It is anticipated that about 250 children exposed to deficient formula and 500 matched control children exposed to other soy formulas will be recruited. In addition, approximately 39 children with a documented history of hypochloremic metabolic alkalosis resulting from defective formula use will be brought to the Washington area. The performance of all these children on a battery of psychological tests will be measured and a careful statistical analysis undertaken to look for an effect of exposure to defective formula with and without documented illness.